Re: fsck, raid reconstruction & bad bad 2.4.3 + some numbers

[Francois Romieu <romieu@cogenit.fr>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=unknown-8bit, Size: 1063 bytes --]

Linas Vepstas <linas@backlot.linas.org> écrit :
[...]
> I want to report a trio of raid-related problems.  The third one is 
> very serious, and effectively prevents 2.4.3 from being usable (by me).
> 
[...]
> The problem:  this dramatically slows fsck after an unclean shut-down.
> You can hear the drives machine-gunning.  I haven't stop-watch timed it,
> but its on the order of 5x slower to fsck a raid partition when there's
> reconstruction going on, then when the raid thinks its clean.  This
> makes unclean reboots quite painful.

Here 2.4.3 takes 15-20 minutes to fsck a 45 Go RAID1 fs after an
unclean shutdown (init=/bin/sh + raid1 autodetect = *boom*). The whole
reconstruction takes an hour (but system is in use).
The disks are IBM-DTLA-307045 on two differents ports of a PIIX4.
The machines includes two 9Go on a 53c875 that perform equally well
(tough a bit noisy during normal startup).

[2.4.3 experience]
> Whatever it is that changed in 2.4.3, it makes unclean reboots
> impossible ...

What does the swap says during the fsck ?

-- 
Ueimor